Inspired by the Japanese vs. American RBI thread and my own boredom, I finally threw a script together to parse the Japanese (Family Stadium) ROM.
http://nightwulf.rbicentral.com/rbi_japan.txt (http://nightwulf.rbicentral.com/rbi_japan.txt)
I plead ignorance on the names. I'm vaguely familiar with Japanese kana, and with the help of a couple websites I wrote a routine to "translate" the player names rather than doing it by hand. I'm fairly confident that they aren't all 100% accurate, but it's the best I can do.
After seeing this interest that the other thread sparked in the Japanese ROM, I'm hoping someone will find this useful. This should also help determine which teams are which, if anyone was working on that.
Nightwulf
So...what would inspire you to "throw together" an editor for RBI 3??? :D :D :D
Quote from: tigermask72 on 05/23/04, 11:48:09 PM
So...what would inspire you to "throw together" an editor for RBI 3??? :D :D :D
Any interest in the game whatsoever.
Nightwulf
Good stuff nightwulf, too tired to really dissect now. But I notice the curve for the text file is using the "old" system instead of being broken down into left-curve and right-curve...
Quote from: Gantry on 05/24/04, 10:25:00 AM
Good stuff nightwulf, too tired to really dissect now. But I notice the curve for the text file is using the "old" system instead of being broken down into left-curve and right-curve...
??? It's really not ... are you looking at curve speed?
Nightwulf
Nice work as usual, Nightwulf. I noticed something interesting - speed values vary more in Family Stadium than in RBI. There's a guy on N with 156 speed, and there are a couple others with 150 speed. There's also a player on N who's slower than Armas, with 116 speed. That team really does stink.
I took a cursory glance at some of the other ratings, and there were a couple things that jumped out: A guy has a 0 contact rating, which is the best possible (Boggs leads RBI with 4), and there's a pitcher with a 16 curve one direction ("Baren", on the M team) - my guess is that either a typo somehow occurred or for some reason the script output values as 1-16 instead of 0-15.
Question - Where is sidearm data stored? Or are there sidearmers in Family Stadium? I haven't played it much.
Quote from: BeefMaster on 05/24/04, 10:37:12 AM
I took a cursory glance at some of the other ratings, and there were a couple things that jumped out: A guy has a 0 contact rating, which is the best possible (Boggs leads RBI with 4), and there's a pitcher with a 16 curve one direction ("Baren", on the M team) - my guess is that either a typo somehow occurred or for some reason the script output values as 1-16 instead of 0-15.
0 contact is definitely valid, as far as what information as to contact's usage I have goes. 16 is also valid for curve ability in either direction, in the Japanese ROM at least. In the USA version, both curve ability attributes are stored together in one byte, giving four bits (0-15) to each. In the Japanese ROM, a full byte (0-255) is alloted to both stats. As other posters have noted, a pitcher with 255 curve ability in the Japanese ROM can "wrap" the ball around the screen several times before reaching the plate.
Quote from: BeefMaster on 05/24/04, 10:37:12 AM
Question - Where is sidearm data stored? Or are there sidearmers in Family Stadium? I haven't played it much.
Under "R/L" in the pitchers, sidearm pitchers are denoted by "LSP" or "RSP" (as opposed to "LHP" or "RHP"). It's stored (almost) identically to the USA version. One byte is used to represent which hand they throw with and whether they throw overhand or sidearm. In the USA version, bits 5-8 of this byte store the "sinker ability" attribute, which has it's own full byte in the Japanese ROM.
Nightwulf
Ah... I hadn't looked closely enough at that part of the text file to see an RSP/LSP - looks like there are several of them.
As for the curve R/L, it maybe provides an insight into why they're combined to one byte in the American version - the programmers might've realized that since half the byte was pretty much never used, they could optimize it down to a single, shared byte. Not that they needed optimizations, though, with fewer teams in the game already. I guess it may have helped when they added 50 teams for RBI 3, if nothing else.
Quote from: BeefMaster on 05/24/04, 12:27:23 PM
As for the curve R/L, it maybe provides an insight into why they're combined to one byte in the American version - the programmers might've realized that since half the byte was pretty much never used, they could optimize it down to a single, shared byte.
In all due respect, I have a better guess. ;)
16 bytes of data are used for each pitcher. In the Japanese ROM, 4 bytes are used for the player name. In the USA version, 6 were used, but there's still exactly 16 bytes of data per pitcher. The extra room came from combining curve left and curve right into one byte, and from combining sinker ability and "R/L handed + SA/OH" into one byte.
My guess is that they combined fields to make room for longer names. Then again, it seems odd that even though a full byte is allocated to both curve abilities and sinker ability in the Japanese ROM, they're still confined to four-bit (0-15) values, with the exception of the pitcher you mentioned above.
Nightwulf
Ah... I didn't know about the name sizes (I know no Japanese, so I hadn't really looked at the names and therefore hadn't taken note of the name lengths). Good call.
My other question - where does the extra ERA info come from? I see values ranging from 1.08 to 4.64. Am I right in guessing that they're using the same single byte as in RBI, but doubling the value added to 1.00 (or not adding it at all, since there aren't any ERAs over 5.11) to come up with the ERA? Your output doesn't have any odd ERAs, which seems to support my guess.
Quote from: BeefMaster on 05/24/04, 01:43:12 PM
My other question - where does the extra ERA info come from? I see values ranging from 1.08 to 4.64. Am I right in guessing that they're using the same single byte as in RBI, but doubling the value added to 1.00 (or not adding it at all, since there aren't any ERAs over 5.11) to come up with the ERA? Your output doesn't have any odd ERAs, which seems to support my guess.
Simpler than that actually. The value in the ROM is multiplied by 4. This gives in-game ERAs a range of 0.00 (0 in ROM) to 9.96 (249 in ROM), in increments of 0.04. Values higher than 249 are not valid, because they would result in ERAs too long to display on the screen. For example, if you use a value of 255 (which should result in an ERA of 10.20), "0.20" is displayed on screen.
Nightwulf
Again, they go with the "Allow way more data than we actually need" method (as I mentioned, they could've handled all the values used in the game using a multiplier of 2)... Maybe the developers intended for the game to be mod-friendly, and they were just ahead of their time.
QuoteIt's really not ... are you looking at curve speed?
Ahh yes... Read, then post. Man, the power numbers for FamSta 87 are really low. I quickly counted
26 guys with a lower power rating than Rafael Santana and way less players in the 800s...